UK.gov to require you to opt out of data-sharing plan with drug firms Plans to share confidential NHS records with private medical researchers have been revealed by Prime Minister David Cameron. The government said in a statement yesterday that it was announcing a consultation "to change the NHS Constitution so that patient data is automatically included in clinical research, but giving patients …
Assuming they can get that right - and it's trustworthy - I can't see a problem with this initiative. At the end of the day, the NHS has paid a lot of money to collect this data ... seems bonkers to sit on it if it can earn some money.
That said, I bet the way it works is the NHS will not see a penny, while a few private companies - that might just have certain ministers as non-exec directors - rake it in.
Sorry, 'who' paid a lot of money? I believe it was the taxpayer (a lot of whom are also patients) who financed this system (which was supposed to be cheaper and more efficient than the mountains of files and paperwork that preceded it). A great idea and may well have saved a lot of lives.
What it was NOT supposed to be, was the basis of another free-for-all for this Governments 'whatever it is - monetise it!' attitude to...well...everything.
As for selling it to private companies on the basis that we get better access to newer treatments? (a justification that was doing the rounds on the BBC News on Monday). Hey guess what, they'll still be extortionately priced (they have to recoup research costs somehow), and if the National Institute for Clinical Excellence decide it doesn't provide value for money?... You simply won't be seeing these new treatments, but thanks for your data anyway. :o)
"Assuming they can get that right"
And that's the worry. Simply taking eg someone's name off the records might not be enough if there are other ways to link patient data together, just as what happened with "anonymised" search details.
But I agree if this can be done then the NHS should be well compensated for the information.
According to a David Willetts on the Today program on Radio 4 yesterday morning, it's already happened in Scotland (about 2:18, but life is too short to fight with the damn awful seek in the BBC's iPlayer). I certainly wasn't asked about this, and will ask my GP when I next see him...
am I not surprised ?
Hands-up all those who remember having to opt-out of all their data being entered into the spineless NHS spine ?
This at a time when people, many of who are in positions of trust such as police and nurses, are being prosecuted/sacked for using their privileged access to data to:
Look-up possible girlfriends
Check-up on boyfriends/girlfriends/wives//husbands/others wives, husbands etc etc etc...
You just CAN NOT trust government at all, never mind with data.
To my gran, who has home visits and been house bound for 8 years to opt out?
her GP has yet to bring the forms for her to view her records as all she gets visited by are locums or the nurse, she hasnt seen her own GP in over 10 years
another NHS failure to consult the people it's shafting err aiding
If the data is anonymised, I don't see any problem. I'm sure there's probably a horrendous number of deaths that result every day from poor treatment combinations, which are never identified because researchers (including government scientists) don't have access to data like this!
What's the point of medical records, if not to improve medical treatment?
It's also reassuring to see the government acknowledging the importance of science for once (pretty much the only area where Britain leads at something).
The best thing for everyone would be to push for thorough identity stripping of the data, and safe, confidential treatment (i.e. don't outsource the preprocessing to street kids in Somalia).
Anonymised data tends to have traces in it that lead back to a person. If this wasn't the case then there would be no context for blood test results and other records and this would be a major obstacle to the data being useful for research.
If this anonymisation were truly possible I'm sure the NHS would be willing to provide a set of anonymised data to test this principle. Want to bet that they won't allow this?
If data is made open, it will do the exact opposite of what you're saying. Instead of enabling big pharma to hide its mistakes, it will allow independent groups to show where problems exist.
Of course, independent studies are already conducted, but no matter how well controlled they are, when cohort sizes are only tens to hundreds of people, there's no way you can properly assess a drug's safety.
But identity can be ever so relevant - location for disease clusters, age and sex for patterns, relationships for genetics, treatment history for complications. Good data is rich and revealing and that sort of quality is opt-in stuff.
If you just take the names out, it's pretty easy to re-nonymise too: how many poorly old grannies are there in my street, well, Mrs Miggins must be [GUID].
My .02, if the anonymous summary data is so useful, open-source it, make it available to us taxpayers to do surprising and incredible things with.
Well, ignoring the concept of "anonymising" data that needs to be specific enough to allow them to process it for useful statistics (which I'm a little skeptical of), this data is being offered to *private* companies.
I'm sure there are some well-meaning, private medicinal research companies out there, but I don't think it would be too unfair that a large number of them will be more than happy to cherry-pick/skew such data in order to bolster their own agenda.
Obviously, this already happens even without this data, but the idea that its being sold will benefit the greater good is (in my cynical mind) unlikely.
but I am still trying to figure out who it is I go to get the payment for each use of my medical record by these private companies. I am going to get paid aren't I??? This is what we are doing this for isn't it, to enable drug companies to develop drugs that make them bigger profits to so they can hand over bigger bonuses to their management?
If the fat cats are getting fatter then I want my cut, no more free data from me. I think a price of £10 per use should see me sorted.
Every treatment you have ever had in your life - many of which have contributed to the likelihood of you living to make this comment - has come from the willingness of people to have their data used by someone that made some money out of it. The NHS has been a major source of information that has improved healthcare across the globe, because it is a centralised, governmental initiative, which makes data-aggregation easy and cheap.
Despite my long-standing aversion to data-creep (look back at previous posts for proof), I cannot see the big problem here - the benefits outweigh the costs, however calculated.
To opt out you simply need to go to the “Government Opt Out Building” where you will need to fill in an “Opt Out” form which you will find in a locked filing cabinet which itself is located in a disused lavatory, the one with a sign on the door that reads “Beware of the Leopard”
This is all of course in the governments “Opt Out Department” which doubles as the Cellar.
Bring your own flashlight.
I thought that too. But sometime we are not in danger of being too cynical ?
The real test is whether the government/NHS can manage to develop a system to produce pure stat-worthy data. I can see why they would need sex, and age. But no need for location data. Instead the requesting company just needs to say "within 30 miles of London", or "at least 10 miles from the nearest connurbation", and the query produces the appropriate records without supplying the location.
I would also expect the system to generate unique IDs for each dataset - although IDs may need to be consistent if you are requesting time-sequenced data, so you know patient X in 2001 is the same patient X in 2011. I don't see why birthdates should be supplied, unless there is weird seasonal-type study going on.
I think most people miss the point of how this data can be used ... statistical analysis of medical data can sometimes reveal incredibly significant, but hitherto unseen links[1]. It is also costly, and therefore not done very often. If by opening up massive tranches of anonymised data to interested companies, we might - just might - push ourselves into the forefront of analysing (and therefore understanding and treating) diseases.
For my own personal situation, MrsJP has multiple sclerosis, and I have long been convinced that when enough data is analysed and cross referenced (in the mother of all Venn diagrams) we will see the "aha !" which will point to a cause - and a cure,
[1]If it were not for an enterprising doctor (Snow) who did this with Insurance company data, in 1854. He overlaid Cholera deaths onto a contour map of London, and was able to tell the government that is was not poverty that carried Cholera (as was thought) - it was water from the Thames. Result ? 30 years before we even knew what Cholera was, London had a world-class sewage system (thanks Bazalgette) and no Cholera.
I've been registered with a doctor once in the last 20 years. It was ten years and three houses ago, for a couple of months in a different county.
My dad was registered. He kept getting pains in his neck. Went to see his GP and was told that he was "quite tall". A few months later died of a massive brain bleed. My wife kept getting out of breath, low iron, bad stomach pains, deadness in the hands. Eventually diagnosed herself as gluten intolerant after years of doctors not knowing what was wrong. Cut out wheat, problem solved. My son has Lyme disease, an ailment about which the NHS has no clue. We paid to go private. I took my daughter into casualty once and they wouldn't see her until I provided the name of her school. (Presumably I was a suspect child abuser).
NHS? Sorry, not a big fan.
it felt like i was stepping on a drawing pin every time i walked. i ended up putting all my weight on the instep. and walking with a slight limp for 2 years.
after 2 years of seeing 3 drs, having an x-ray and being put on tablets etc i asked them to be referred to a podiatrist. i got in within a week.
the podiatrist told me i had a corn. she got a scalpel and cut it off. all i do now when it plays up is grab a stanley knife and do the same.
my father who had Parkinson's disease was told to his face that there was nothing wrong with him and that maybe he was overreacting.
my wife who was due to give birth in 6 weeks was told that she was ok, even after all the symptoms for pre-eclampsia were showing. the midwife said not to come back for a month as she was on holiday.
luckily her mum was an ex-midwife and took her to hospital where she was rushed in for a c-section less than 24 hours later. she only just survived and our son would have definitely died. he was 6 weeks early but very healthy and didnt need to even go in the baby care unit.
yeah, i trust the Drs & NHS a lot! :(
I campaign on medical privacy and to me it looks like you will be of Researchers contacting you but be able to opt out of your medical data which will be anonymised down to date of birth and postcode. See Prof Ross Anderson Lightbluetouch paper blog for the issues of anonymised data.
I can see in a few years we will be opting in to view adult sites BUT opting out of our personal data being sold to the highest bidder. anonymised data is easy to piece back together remember the AOL search data that was "anonymised" but was easy enough to put everything together and come up with a name.
For that that think anonymised data is safe, do a Google search on Mass. Governor Weld. Mass. did a similar thing and a journalist was able to get Weld's medical data with a few simple searches. If you know a few details about someone, such as gender, age, specific medical conditions(diabetes, heart disease, epilepsy, etc) as well as even a general location, you may be able to generate a complete profile of individuals and discover what they have been hiding, such as HIV status, anti-depressant use, STDs, and so forth. I am sure that no one will mind all this information being sent to pharma companies, and all the outsourcing companies that they use./sarcasm
Write to your MP, tell them to make sure this doesn't happen. Tell them that Tories and Lib Dems won't see power for a generation if they go through with it.
Medical information is the *MOST* sensitive data stored. Nobody should get hold of it except qualified medical professionals, and only in circumstances directly pertinent to your health.
It's surprisingly easy to reassemble disaparate information to find out who the original person is. Ever heard of web profiling?
Anyway, the point is somewhat missed - the government doesn't "own" this information, it is given to medical staff in confidence for the purpose of diagnosis. I can't imagine that this idea would pass data-protection requirements.
If you opt out, then the govt/nhs have to keep your records in a separate, secure confidential database, away from the more readily available one.
So you're guaranteed that someone will leave a copy of your health records unlocked on a laptop or usb drive on a train or in a taxi every 4 months. Whereas, by staying opted in, your data should be safer by obscurity, as it'll be lost amongst all the rest of the populations data which will be held in a privately run database with the correct levels of access controls and redundant backup systems.
Possibly, they've spent the last 5 years consistently "loosing" people's private health data in order to convince us that being auto-opted-in really is the better and safer choice.
You're really not getting it are you? The information the NHS hold about you is not _their_ data... it's _your_ data. They have already undertaken a commitment to hold it on your behalf with your permission, and they have to ask before they share it with anyone because it doesn't belong to them!
There is no circumstance in which the NHS has the right to profit from the sale of _your_ information, unless you give them permission to of course. _That_ is why this process is happening.
But instead of asking if it's OK to profit from your personal information, they've said "we're going to sell something that belongs to you. Here's a bureaucratic process to follow if you don't like it" - does that sound like an open and protective government to you? Or does it rather sound like a load of shysters working to help their corporate chums get stinking rich?
"opting NHS patients into its data-sharing plans would help remove delays within the system"
This really is double-speak at its worst: "Opt: To make a choice or decision from a range of possibilities"
If the proposed system had any credibility they'd have the courage of conviction to rely on opt-in on a per-patient basis.
What's proposed instead is that all data is stolen and re-sold, with patients later given some (most likely obscure) mechanism to opt out. It's clearly not compatible with either the spirit or the letter of DP laws to use people's medical data in this way.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
